What are the nature of the problems?

00:00

Three categories of acid-base / electrolyte problems with CRRT
Given this background for you, what do you expect to find as complications with CRRT? They could be secondary to the patient condition itself. They could be contributed to by the therapy. Or they are purely iatrogenic because of different issues.

00:00

Secondary to patient condition
In terms of problems secondary to the patient condition itself, severe metabolic acidosis, the results of permissive hypercapnia, as Dr. Luce has already shown you, alkalosis, hypernatremia, hyperphosphatemia are common in ICU patients. Often times because of the nature of CRRT patients, it is hard to tell--was this a primary problem or was this aggravated by the use of CRRT?

00:00

Case study: 54 year old patient after a lung transplant
We have successfully used citrate CRRT actually to treat permissive hypercapnia because it gives you an opportunity to adjust the base and the fluid balance very independently. This is an example of a patient who had a single lung transplant. On Day #4 following his transplant, he extubated himself, was fatigued, was reintubated, had worsening hemodynamic and respiratory parameters.

By the time we saw him, he was on pressors, he had a cardiac output of 6.8, he had a low vascular resistance, and his lungs had coarse rales, but he was on two ventilators with 100 percent FIO₂ on each lung. Notice that he was still making some urine on lasix, but his blood urea nitrogen level was elevated. Most importantly, his bicarbonate was 31, anion gap of 10, but his pH was 7.16, and a PCO₂ was 87. His oxygen saturation despite this regimen was only 89.5 percent. This is a typical scenario where you are forced to trade off permissive hypercapnia to try and reduce the lung injury. The trouble here with the intensivists asking us was... could we help them with the fluid management and the acid base?

00:00

We started him on CVVHDF using roughly the same parameters I just showed you and same composition. The only thing we did was instead of using 23 percent saline to make up the final concentration of sodium chloride, we substituted it with bicarbonate. So the final sodium was still 117. As a consequence, you can see we were able to deliver bicarbonate back. Over the next 24 hours, his acidemia improved.

00:00

Laboratory values during the patient's course
I will show you that graphically here. Here 16 hours before we start. This is four hours after. This is the CO₂. It doesn't change a whole lot. But what you can see is that the amount of bicarbonate being given is reduced, and his pH gradually tends to improve over time. His PCO₂ requirement comes down as do his FIO₂ requirements. All we did was support him. I don't thin we took any CO₂ off via the filter. I don't think we did anything of that nature. But we were able to regulate his fluid better and provide him buffer without fluid overloading him. So that is what I mean that you have the ability to manipulate (that).

00:00

Acid-base / elecrolyte problems worsened or caused by therapy
Now what about the acid-base problems that are contributed by therapy? If you are able to do it in this manner, you have the opportunity then to influence acid-base balance, electrolytes, the glucose, phosphate, and also can result in some other specific problems related to citrate and fluid use.

00:00

In terms of the acid-base imbalance, the issue is increased base loss. Remember when you put these people on therapy, you have 24 ml/minute of lactate clearance. The bicarbonate sieving coefficient is about 1.12. So you will remove bicarbonate whether or not you replace it. The question then becomes: How much do you replace? If you don't replace it often enough, you will have an increased base loss. You can overcorrect it and get overshoot alkalosis. And if you give too much, you can have some CO₂ retention.

00:00

Serum bicarbonate changes in Case No. 1
This is the same patient I started off with, the patient with the coronary bypass graft. As you can see here, this is the bicarbonate change. This is the pH change. We start here at time zero. You find that her bicarbonate rapidly comes down. Over a period of about 12 to 16 hours, it is so low we are somewhat puzzled. You see the pH comes down a lot. This is not what we expect because what we had started her on was citrate, her starting bicarbonate had been about 22. We didn't anticipate it to be coming down this rapidly. We explored this. What had ended up happening was we had prescribed citrate at 180 ml/hour. But the nurse who took off the order put it at 80 ml/hour. So now you can see I am taking more bicarbonate off. I don't have enough delivery of citrate as a base.

00:00

By the time this is recognized, about 12 to 15 hours later, you have to give more bicarbonate, change the citrate back and then you can achieve steady state again. So you can see that the pH starts improving, but this is purely a therapy- delivered issue which is iatrogenic in the sense we made a prescription but it was not followed. Now you could take it to the point that if you had not anticipated this, you would have expected some of these things to happen.

00:00

Hypo- and hypernatremia
What about hypo- and hypernatremia?

00:00

Source: Locatelli F et al, Kidney Int Suppl. 1998 May;66:S151-5. Review.

Factors affecting sodium removal during dialysis
I think it is important to recognize that when you are dealing with sodium concentrations--this is an article by Dr. Locatelli from Vincenza. There is a difference in terms of plasma water because you have about 7 percent of plasma as protein. So sodium is distributed only in the water. But when we measure it, we measure it across and we reflect it as it is across the entire, 100 percent, of fluid. In the ultrafiltrate, you will expect to find only the plasma water component of sodium coming across.

00:00

Source: Locatelli F et al, Kidney Int Suppl. 1998 May;66:S151-5. Review.

Sodium removal with different types of CRRT
If you were to look at different techniques, and this is assuming that you have an ultrafiltration rate of 8 ml/min in the CAVH technique, which was simulated here, an equivalent you could do with CAVH, and dialysate of 1 liter an hour, you can see that at the three different starting plasma compositions, you are likely to remove sodium if you keep your substitution fluid constant at 140.5 and your dialysate at 132. These are the different parameters. So in general, you have extreme ability to remove sodium. However, depending upon how you manipulate these fluids, you can also cause hypernatremia.

00:00

Serum sodium and potassium levels in Case No. 1
This is again the same example of the patient we started with. You can see that the starting sodium was fine. It was being maintained okay. But because of the bicarbonate load, which resulted from the acidemia which occurred, you can see he started getting a little hypernatremic until that is corrected back. And over time, you can see that that changes. This is potassium. It starts coming down, as you see that.

00:00

Hyperglycemia
Another aspect of this hyperglycemia.

00:00

Recognize that we are very used to in dialysis thinking about only removing things. However, you have the ability to deliver molecules across the other direction.

00:00

Source: Sigler MH and Teehan BP. Kidney Int. 1987 Oct;32(4):562-71.
Monaghan R et al, Crit Care Med. 1993 Aug;21(8):1159-63.

Glucose transfer in IHD vs. CRRT
In most dialysis circuits, we only use glucose of 0 to 100 at the most. However, when you are using Dianeal solutions, which have 1.5 grams percent of dextrose and you are providing it continuously, you are going to have dextrose transfer across into the blood. That is what causes hyperglycemia. So if you were to quantify this, if you use a 1.5 percent dextrose solution, you have a 43 percent uptake of glucose, which will deliver about 525 kilocalories per day. If you use a 2.5 percent dextrose, you will deliver almost 1,000 calories a day.

This is reflective of the patient who has dextrose in the dialysate. It is not uncommon for you to use D5W with bicarbonate at a substitution fluid. If you do that, you are delivering the same amount on the other side directly into the blood stream. So this problem can occur both in hemofiltration and in hemodialysis.

00:00

Source: Monaghan R et al, Crit Care Med. 1993 Aug;21(8):1159-63.

Glucose uptake during CVVH depending on replacement fluid (D5 plus electrolytes) infusion rate
This is just to show you that in hemofiltration if you are using a D5 solution, you will see the rate of infusion here, the glucose uptake grams per hour.

00:00

Case study: Hyperglycemia due to glucose in CVV-HDF fluids
Why is this important to recognize? As shown in this patient example here, this was a lady who was 62-years old with end-stage liver disease. She was awaiting liver transplantation, and this was at a time when we were using CVVHDF D5 with three amps of bicarbonate. She was on 2 percent dextrose in addition for her tube feedings, and notice her blood sugars go up there.

00:00

Caloric intake from CRRT- and infusion-related glucose
But if you compute her caloric intake, you would have expected her calorie goals to be only about 2,000 a day. But because of all the dextrose she is getting, she actually ended up getting about 5,000 calories a day. That is an enormous nutritional load for her to deal with. That is what contributes to the hyperglycemia and problems in this situation.

00:00

Hypophosphatemia
What about hypophosphatemia?

00:00

Serum phosphate levels in Case No. 1
If you notice, this is the same patient I started off with. These are her phosphate levels. Predictably she came in with a low level, which I think was erroneous because four hours later it was 3.8. You can see that right over a period of about 30 hours, she comes down as low as 1.3. This is because we do not have any phosphate in the dialysate.

00:00

Phosphorus kinetics in CRRT
We anticipate this for the following reasons. We had done a study some years ago which had shown that actually in contrast to intermittent hemodialysis where phosphate is a two-compartment model, in continuous therapies it can be predicted by single pool kinetics. And what you expect to happen... the changes in serum phosphate are because you are clearing it rapidly across the filter, but there are also predominant and significant intracellular shifts. These are contributed to by the glucose reabsorption and also refeeding.

Most of these patients have not had adequate nutrition for awhile. When you start feeding them, you push phosphate back into the cells, and you come up with refeeding syndrome. What you can expect to happen is if you have a 70 kg man starting with a serum phosphate of 6 and he's on a CVVHDF of 1 liter dialysate an hour and UFR of 600, you would expect the phosphate down to about 3 in about 30 hours.

00:00

Risk factors for hypophosphatemia
Here is the at-risk population that you should worry about particularly with this. These are patients with chronic malnutrition, acute malnutrition. These are patients in different situations who been NPO 7 to 10 days or have diabetic ketoacidosis or CRRT. These are the main things in terms of refeeding.

00:00

Refeeding syndrome
Just to give you an example of what the refeeding actually entails. In starvation there is loss of lean muscle mass, water and minerals. Serum phosphate is within normal limits. However, as you give these patients carbohydrates, there is insulin release, there is enhanced cellular uptake, and the total body phosphate is depleted with increased cellular uptake, resulting in hypophosphatemia. And this is a significant factor to consider when you've got prolonged respiratory dysfunction because it will influence your ability for muscle contraction and worsen that aspect of it.

To go to PART THREE click here

Back to Topic Index
Topic Index  Chronic Renal Failure -- Continuous Renal Replacement Therapies